This invention relates to semiconductor fabrication technology and, more particularly, to liquid phase epitaxial growth techniques.
One important application of silicon charge transfer devices is as imagers in the visible region of the spectrum. II-VI compound semiconductors can be used in a similar manner to extend the imaging capability to the infrared region, thereby rendering the latter devices particularly important for national defense applications requiring the detection of infrared radiation. The Hg.sub.1-x Cd.sub.x Te alloy, for example, is an intrinsic II-VI compound semiconductor whose alloy content can be adjusted to cover any part of the spectral range from 0.8 to over 30.mu.m. HgCdTe can be utilized in an imaging function in a hybrid focal plane, wherein an HgCdTe detector is mated to a silicon signal multiplexer. An even more desirable imaging arrangement is realized in a monolithic focal plane, which is formed from a single semiconductor. In this design, the semiconductor includes an HgCdTe sensor layer for photon detection, while a CdTe layer is provided for signal processing functions. The CdTe layer exhibits a wide bandgap, so that the dark current therein is inherently low. Furthermore, this semiconductor is intrinsic by nature, and thereby operates at a higher temperature than extrinsic silicon, permitting the realization of a heterojunction structure in the detector. In this manner, the monolithic focal plane technique makes possible the detection of low infrared energy in a narrow gap semiconductor and the transfer of the resulting charges to a wider gap semiconductor for singal processing.
Although II-VI semiconductor compounds are required for these applications, attempts at producing high quality II-VI compounds have met with only limited success in the prior art. Production has been restricted for the most part to the bulk type of crystals, which are obtained by solidification from a melt. Unfortunately, however, bulk crystals, when used in an imaging application, are restricted to frontside illuminated modes, unless the crystals are backside thinned. In addition to this deficiency, there is a need for a production technique which will yield large area, uniform Hg.sub.1-x Cd.sub.x Te layers. Such a material would exhibit a greater efficiency and a higher operating temperature than the materials which previously have been obtainable.
One way in which these needs could be met is with an epitaxially grown material. An epitaxial imager, for example, is capable of operating in a backside illuminated mode without substrate thinning. This feature would render the epitaxially grown devices more compatible with the hybrid focal plane configuration. Furthermore, the lower growth temperature which is inherent in liquid phase epitaxial techniques, as compared with bulk growth, would yield compounds of a more uniform composition. Finally, a relatively thin epitaxial layer, if such a layer could be grown, would limit the bulk generation volume of a diffusion current within the material. As a consequence, epitaxial diodes fabricated from such material would be expected to have a higher R.sub.o A product.
Although the liquid phase epitaxial growth method would thus provide a higher quality, more useful product, this technique has not been heretofore successfully applied to the growth of II-VI compounds. Liquid phase epitaxy, which is a relatively low temperature growth process developed extensively in connection with the preparation of high quality III-V and IV-VI semiconductors, could be used to solve two major problems encountered in II-VI compound bulk crystal growth, the compositional nonuniformity which is experienced and the long annealing times which are necessary to reach homogeneity in the bulk materials. Only limited attempts have been made, however, to apply liquid phase epitaxial techniques to the growth of II-VI compounds. The high vapor pressures characteristic of Column II elements make it difficult to maintain the proper concentration of the Column II elements within the growth solution during growth. In addition, liquid phase epitaxial techniques are further limited by the low solubility of Column VI elements in Column II elements at the relatively low temperatures used in the epitaxial growth techniques.
Therefore, a need has developed in the art for a liquid phase epitaxial growth system which may be utilized to grow high quality II-VI compound semiconductors.